Effect of a web-based curriculum on primary care practice: basic skin cancer triage trial

Primary care physicians (PCPs) are uniquely positioned to detect melanoma. Effective educational interventions targeted at PCPs may improve early melanoma detection. A previous in-person Basic Skin Cancer Triage (BSCT) 2-hour course demonstrated significant short-term improvement in provider practices, attitudes, ability, confidence, and knowledge. We conducted a randomized trial to test the efficacy of the BSCT course implemented as a web-based learning program, compared to a similar (control) web-based course on weight assessment.

We recruited a sample of 57 PCPs and 3,341 of their patients from four geographically diverse centers. Skin cancer control activities by PCPs were assessed by physician survey and by chart review and patient telephone interview about their recent visit to their PCP at baseline and at 1--2 months and 12 months after course completion.

Some effect of intervention on skin cancer parameters was self-reported by physicians; this was not confirmed by patient survey or chart-extracted data. Rates of skin cancer control practices by PCPs were low across both groups before and after intervention. The positive changes in physician-reported behaviors (total body skin examination [TBSE]), intentions (discuss skin cancer detection), confidence (performing TBSE), office practices, and knowledge (58% skin versus 49% control) were neither matched by differences in practice reported by their patients, nor persisted in a longer term follow-up, hence may be attributable to physician recall bias due to the experience of the course or desire to please study investigators and were less dramatic as compared to our previously reported in-person BSCT intervention. Thus this approach by itself appears unlikely to result in improved PCP handling of skin cancer issues.

Given previous success with our in-person course, the features required to make WBL a more effective tool for medical education must be further explored.